Controle de um reator tubular encamisado por meio de curvas de operação geradas online por computação paralela

Abstract

Currently, there is a growing demand for faster and more efficient data processing, a demand that also affects computational simulations. To meet this need, an alternative is the use of parallel computing in program development. Among the different tools that allow the application of parallel computing, one of the best known is the CUDA platform, which allows the use of NVIDIA graphics cards for data processing. To test the effect of parallel computing on the performance of a computational simulation, a Python simulation of a jacketed tubular reactor was developed, and this was written and executed in both serial and parallel forms, using CUDA and in the cloud using the Google Colab platform. The results obtained show a great advantage in using graphics cards for parallel processing in cases where there is a large volume of data to be processed, and in cases where it is possible to parallelize the program. In cases where a program is very extensive, without the possibility of parallelization, normal processors still show better times